adding broadcast where operator

backends/cadence/hifi/kernels/CMakeLists.txt

@@ -9,6 +9,7 @@ add_library(
   cadence_kernels
   kernels.cpp
   ${EXECUTORCH_ROOT}/backends/cadence/hifi/third-party/nnlib/matmul_asym8uxasym8u_asym8u.cpp
+  ${EXECUTORCH_ROOT}/backends/cadence/hifi/third-party/nnlib/xa_nn_elm_where_f32xf32_f32.c
   ${EXECUTORCH_ROOT}/backends/cadence/hifi/third-party/nnlib/xa_nn_elm_div_broadcast_f32.c
   ${EXECUTORCH_ROOT}/backends/cadence/hifi/third-party/nnlib/xa_nn_elm_floor_div_broadcast_f32.c
 )

backends/cadence/hifi/kernels/kernels.h

@@ -15,6 +15,7 @@
 /* For NNLIB APIs */
 #include "xa_nnlib_kernels_api.h"
 
+
 extern "C" WORD32 xa_nn_elm_floor_div_f32xf32_f32(
                                     FLOAT32 * __restrict__ p_out,
                                     const FLOAT32 * __restrict__ p_inp1,
@@ -36,7 +37,17 @@ extern "C" WORD32 xa_nn_elm_div_broadcast_4D_f32xf32_f32(
                                     const WORD32 *const p_inp1_shape,
                                     const FLOAT32 * __restrict__ p_inp2,
                                     const WORD32 *const p_inp2_shape);
-
+
+extern "C" WORD32 xa_nn_elm_where_f32xf32_f32(FLOAT32 * __restrict__ p_out, const FLOAT32 * __restrict__ p_inp1, const FLOAT32 * __restrict__ p_inp2, const unsigned char *__restrict__ p_condition, WORD32 num_elm);
+
+extern "C" WORD32 xa_nn_elm_where_broadcast_4D_f32xf32_f32(FLOAT32 * __restrict__ p_out, const WORD32 *const p_out_shape,
+                      const FLOAT32 * __restrict__ p_inp1,
+                      const WORD32 *const p_inp1_shape,
+                      const FLOAT32 * __restrict__ p_inp2,
+                      const WORD32 *const p_inp2_shape,
+                      const unsigned char *__restrict__ p_condition,
+                      const WORD32 *const p_condition_shape
+                      );
 
 namespace impl {
 namespace HiFi {

backends/cadence/hifi/operators/CMakeLists.txt

@@ -23,6 +23,7 @@ set(_aten_ops__srcs
     "${EXECUTORCH_ROOT}/kernels/portable/cpu/op_add.cpp"
     "${EXECUTORCH_ROOT}/kernels/portable/cpu/op_embedding.cpp"
     "${EXECUTORCH_ROOT}/kernels/portable/cpu/op_full.cpp"
+    "${EXECUTORCH_ROOT}/backends/cadence/hifi/operators/op_where.cpp"
     "${EXECUTORCH_ROOT}/kernels/portable/cpu/op_view_copy.cpp"
     "${EXECUTORCH_ROOT}/kernels/portable/cpu/util/activation_ops_util.cpp"
     "${EXECUTORCH_ROOT}/kernels/portable/cpu/util/copy_ops_util.cpp"

backends/cadence/hifi/operators/op_where.cpp

@@ -0,0 +1,150 @@
+/*
+ * Copyright (c) Meta Platforms, Inc. and affiliates.
+ * All rights reserved.
+ *
+ * This source code is licensed under the BSD-style license found in the
+ * LICENSE file in the root directory of this source tree.
+ */
+
+#include <executorch/kernels/portable/cpu/util/broadcast_util.h>
+#include <executorch/kernels/portable/cpu/util/functional_util.h>
+#include <executorch/runtime/kernel/kernel_includes.h>
+#include "kernels.h"
+
+namespace torch {
+namespace executor {
+namespace native {
+
+#define NNLIB_MAX_DIM 4
+
+Tensor& where_out(
+    RuntimeContext& ctx,
+    const Tensor& cond,
+    const Tensor& a,
+    const Tensor& b,
+    Tensor& out) {
+  ScalarType cond_type = cond.scalar_type();
+  ScalarType a_type = a.scalar_type();
+  ScalarType b_type = b.scalar_type();
+  ScalarType common_type = promoteTypes(a_type, b_type);
+  ScalarType out_type = out.scalar_type();
+
+  ET_KERNEL_CHECK(ctx, common_type == out_type, InvalidArgument, out);
+
+  // Determine output size and resize for dynamic shapes
+  ET_KERNEL_CHECK(
+      ctx,
+      resize_to_broadcast_target_size(a, b, cond, out) == Error::Ok,
+      InvalidArgument,
+      out);
+
+  constexpr auto name = "where.self_out";
+
+  ET_CHECK_MSG(
+      cond_type == ScalarType::Bool || cond_type == ScalarType::Byte,
+      "Unhandled dtype %s for where.self_out",
+      torch::executor::toString(cond_type));
+
+      /*logic to find broadcast*/
+      const int a_is_broadcasted = !out.sizes().equals(a.sizes());
+      const int b_is_broadcasted = !out.sizes().equals(b.sizes());
+      const int cond_is_broadcasted = !out.sizes().equals(cond.sizes());
+      const int broadcast = (a_is_broadcasted || b_is_broadcasted || cond_is_broadcasted);
+
+      int max_dim = a.dim() > b.dim() ? a.dim() : b.dim();
+      max_dim = out.dim() > max_dim ? out.dim() : max_dim;
+      bool fall_back = 0;
+      if((a_type != ScalarType::Float) || (b_type != ScalarType::Float))
+        fall_back = 1;
+
+      if((broadcast == 1) && (max_dim > NNLIB_MAX_DIM))
+          fall_back = 1;
+
+if(!fall_back)
+{
+      const float* a_data = a.const_data_ptr<float>();
+      const float* b_data = b.const_data_ptr<float>();
+      float* out_data = out.mutable_data_ptr<float>();
+      const unsigned char* con = cond.const_data_ptr<uint8_t>();
+
+      if(broadcast == 1)
+      {
+         int out_shape[NNLIB_MAX_DIM];
+         int inp1_shape[NNLIB_MAX_DIM];
+         int inp2_shape[NNLIB_MAX_DIM];
+         int con_shape[NNLIB_MAX_DIM];
+
+         for(int i = 0; i < NNLIB_MAX_DIM; i++)
+         {
+            con_shape[i] = 1;
+            out_shape[i] = 1;
+            inp1_shape[i] = 1;
+            inp2_shape[i] = 1;
+         }
+
+         int off_o = NNLIB_MAX_DIM - out.dim();
+         int off_a = NNLIB_MAX_DIM - a.dim();
+         int off_b = NNLIB_MAX_DIM - b.dim();
+         int off_c = NNLIB_MAX_DIM - cond.dim();
+
+         for(int i = 0; i < out.dim(); i++){
+            out_shape[i+off_o] = out.size(i);}
+
+         for(int i = 0; i < a.dim(); i++)
+            inp1_shape[i+off_a] = a.size(i);
+         for(int i = 0; i < b.dim(); i++)
+            inp2_shape[i+off_b] = b.size(i);
+         for(int i = 0; i < cond.dim(); i++)
+            con_shape[i+off_c] = cond.size(i);
+
+        /* Add fallback if broadcast and condition dimension are larger than inputs dimension */
+
+        if(con_shape[0] != out_shape[0] || con_shape[1] != out_shape[1] || con_shape[2] != out_shape[2] || con_shape[3] != out_shape[3])
+        {
+            void* p_scratch =  malloc(out_shape[0]*out_shape[1]*out_shape[2]*out_shape[3]);
+            const unsigned char *p_brd_cond = (const unsigned char*)p_scratch;
+            xa_nn_broadcast_8_8((WORD8* __restrict__) p_brd_cond, out_shape, (const  WORD8* __restrict__) con, con_shape, 4);
+
+            for(int i = 0; i < 4; i++)
+            {
+                con_shape[i] = out_shape[i];
+            }
+            xa_nn_elm_where_broadcast_4D_f32xf32_f32(out_data, out_shape, a_data, inp1_shape,
+                                                    b_data, inp2_shape, p_brd_cond, con_shape);
+            free(p_scratch);
+        }
+        else
+        {
+            xa_nn_elm_where_broadcast_4D_f32xf32_f32(out_data, out_shape, a_data, inp1_shape, b_data, inp2_shape, con, con_shape);
+        }
+      }
+      else
+      {
+          xa_nn_elm_where_f32xf32_f32(out_data, a_data, b_data, con, out.numel());
+      }
+}
+else
+{
+  ET_SWITCH_REALHB_TYPES(a_type, ctx, name, CTYPE_A, [&]() {
+    ET_SWITCH_REALHB_TYPES(b_type, ctx, name, CTYPE_B, [&]() {
+      using CTYPE_OUT =
+          typename torch::executor::promote_types<CTYPE_A, CTYPE_B>::type;
+      apply_ternary_elementwise_fn<CTYPE_A, CTYPE_B, uint8_t, CTYPE_OUT>(
+          [](const CTYPE_A val_a, const CTYPE_B val_b, const uint8_t val_c) {
+            CTYPE_OUT a_casted = static_cast<CTYPE_OUT>(val_a);
+            CTYPE_OUT b_casted = static_cast<CTYPE_OUT>(val_b);
+            return val_c ? a_casted : b_casted;
+          },
+          a,
+          b,
+          cond,
+          out);
+    });
+  });
+}
+  return out;
+}
+
+} // namespace native
+} // namespace executor
+} // namespace torch